Fig 1.
Loss of NS1-BP protein inhibits viral M mRNA nuclear export without significantly altering intracellular distribution of bulk cellular poly(A) RNA.
(A) Disruption of the NS1-BP gene by CRISPR-Cas9 system yielded A549 cells lacking NS1-BP protein. Cell lysates from control or NS1-BP knockout cells were subjected to western blot analysis with antibodies against NS1-BP antibody or β-tubulin, as control. (B) Wild-type or NS1-BP-/- A549 cells were infected with influenza virus (A/WSN/33) at MOI 2 for 6h. Single-molecule RNA fluorescence in situ hybridization (smFISH) was performed to detect influenza virus M mRNA. Hoechst staining labeled nuclei. Scale bar = 10 μm. (C,D) Quantification of total fluorescence intensity of M mRNA in the nucleus and cytoplasm of wild-type or NS1-BP-/- cells (C), or nuclear-to-cytoplasmic (N/C) ratios of M mRNA in these cells (D) from panel B. (E) Non-infected wild-type or NS1-BP-/- A549 cells were subjected to RNA-FISH to label poly(A) RNA. (F,G) Quantification of total fluorescence intensity of poly(A) RNA in the nucleus and cytoplasm of wild-type or NS1-BP-/- cells (F), or nuclear-to-cytoplasmic (N/C) ratios of poly(A) RNA in these cells (G) from panel E. Thirty cells were quantified for each condition. These data are representative of three independent experiments. ***p<0.001.
Fig 2.
Schematic representation of a high-throughput screen to identify chemical inhibitors of viral M mRNA processing and nuclear export.
Screen was performed using a chemical library of 232,500 compounds in A549 cells. Cells were incubated with compounds for 30 min and, for robust imaging analysis, ~ 100% of the cells were infected with influenza virus (WSN), at MOI 2 for 7.5h. Viral M mRNA was detected by smRNA-FISH and images were systematically taken in a high throughput microscope (IN Cell Analyzer 6000). Samples on 384-well black clear-bottom plates were imaged at 20X magnification using the Hoechst and dsRed filters. 4 fields of view per well were collected for each channel. The distribution of fluorescent signal between the nucleus (N) and the cytoplasm (N/C ratio) as well as total cell signal intensity were quantified using GE IN Cell Analyzer Workstation (version 3.7.3) and Pipeline Pilot (version 9.5; Biovia). Data was imported into the GeneData’s Screener™ software analysis suite for quality control to ensure that data quality is high for all plates in each experimental run (Z’ > 0.4).
Fig 3.
Identification of small molecules that inhibit viral M mRNA nuclear export and/or decrease viral M mRNA levels.
(A) Representative images showing uninfected cells; cells infected with A/WSN/33 and pretreated with 0.5% DMSO (control), which show viral M mRNA exported into the cytoplasm, in red; and cells infected with A/WSN/33 pre-treated with 2.5 μM of the transcription inhibitor DRB (5,6-dichloro-1-β-D-ribofuranosylbenzimidazole). DRB served as positive control for viral M mRNA nuclear retention. (B) Distribution of nuclear (N) to cytosolic (C) (N/C ratio) of all wells in a mock assay plate showing the assay window and sensitivity. DMSO wells were normalized to 0 and DRB positive control wells were normalized to 100. The circled red diamond shape represents a lower dose of DRB and shows lower normalized N/C ratio than the other diamonds representing the full control dose of DRB. (C) Rank-sorted Z-score of the Nuclear to Cytoplasmic (N/C) ratio of viral M mRNA in A549 cells after individual treatment with 232,500 compounds (2.5 μM). Each N/C value is expressed as a Z-score, indicating the number of standard deviations from the median plate ratio. Points above the red line at Z-score 3 represent compounds that were considered hits in the primary screen. (D) Rank-sorted Z-score of the total intensity of viral M mRNA after compound treatment. Each value is expressed as a Z-score, indicating the number of standard deviations from the median plate intensity. Points below the red line at Z-score -3 represent compounds selected as hits in the primary screen. To better visualize the distribution of compounds within the desired range (Z-score < -3), the Z-score range of the graph has been focused to view data points that show decrease in viral M mRNA fluorescence.
Fig 4.
Schematic representation of identification and selection of top hits that inhibit viral M mRNA nuclear export and/or expression.
Out of the 232,500 compounds tested in the primary screen, we selected compounds with Z-scores ≥ 3 for the N/C ratio and compounds that decreased viral mRNA levels with Z-scores ≤ -3. Compounds that reduced nuclear count significantly (Z-score < -3) were considered cytotoxic and were eliminated from further consideration. Of those remaining, the 1,125 compounds with the highest Z-scores were chosen for confirmation studies. The top 600 compounds from single-dose confirmation studies were further evaluated in a 12-point dose response study to assess the potency (AC50 –concentration at 50% activity). Examples of dose-response curves showing phenotypes of hits that induced viral M mRNA nuclear export block (increased N/C) and decrease in viral M mRNA levels (decreased intensity) are depicted. During this step, bulk cellular poly(A) RNA localization and intensity were also assessed by smRNA-FISH to determine the effect of these compounds on host cell mRNA (intensity and N/C ratio). Compounds that inhibited viral mRNA nuclear export and/or decreased viral mRNA levels but had no significant effect on the host cell poly(A) RNA were then selected for additional assays.
Fig 5.
Compound 2 inhibits viral mRNA nuclear export.
(A) Structure of compound 2. (B) RNA-FISH and smRNA-FISH followed by fluorescence microscopy were performed in cells treated with 0.1% DMSO or 2.5 μM compound 2 to detect poly(A) RNA and GAPDH mRNA, respectively, in uninfected cells. (C-F) Total fluorescence intensity or nuclear to cytoplasmic fluorescence intensity (N/C ratio) were quantified for poly(A) RNA and GAPDH mRNA in the absence or presence of compound 2. For C (C, n = 174 cells; Compound 2, n = 181), D (C, n = 172 cells; Compound 2, n = 181 cells), E (C, n = 166 cells; Compound 2, n = 181 cells), and F (C, n = 151 cells; Compound 2, n = 160 cells). (G) Cells were treated as in B except that smRNA-FISH was performed with probes to detect M mRNA in cells infected with WSN at MOI 2 for 8 h. (H,I) Total fluorescence intensity or nuclear to cytoplasmic fluorescence intensity (N/C ratio) were quantified for M mRNA in the absence or presence of compound 2. For H (C, n = 91 cells; Compound 2, n = 104 cells) and I (C, n = 101 cells; Compound 2, n = 95 cells). (J) Cells were treated as in G except that smRNA-FISH was performed with probes to detect HA mRNA. (K,L) Total fluorescence intensity or nuclear to cytoplasmic fluorescence intensity (N/C ratio) were quantified for HA mRNA in the absence or presence of compound 2. For K (C, n = 104 cells; Compound 2, n = 137 cells) and L (C, n = 101 cells; Compound 2, n = 126 cells). (M) Cells were treated as in G except that smRNA-FISH was performed with probes to detect NS mRNA. (N,O) Total fluorescence intensity or nuclear to cytoplasmic fluorescence intensity (N/C ratio) were quantified for M mRNA in the absence or presence of compound 2. For N (C, n = 96 cells; Compound 2, n = 135 cells), and O (C, n = 106 cells; Compound 2, n = 113 cells). At least three independent experiments were performed for each imaging analysis. (P,Q) Relative mRNA ratios of M2 to M1 (P) and NS2 to NS1 (Q) were determined by qPCR from RNA obtained from cells infected as in G and treated with 0.1% DMSO, 1 μM, or 2.5 μM compound 2. The nuclear speckle assembly factor SON was knocked down with siRNAs as positive control for inhibition of M1 to M2 mRNA splicing. Three independent experiments were performed. C, control. (R) Cellular ATP levels were measured in cells treated with 0.1% DMSO or 2.5 μM of compound 2 at 24 h. Four independent experiments were performed and each contained 6 technical replicates. Graphs shows data points and mean +/- SD. *p<0.05; ***p<0.001, ****p<0.0001.
Fig 6.
Partial depletion of the mRNA export factor UAP56 show differential export of viral mRNAs similar to compound 2.
(A) smRNA-FISH followed by fluorescence microscopy was performed to detect M mRNA in A549 cells treated with control siRNA or with two concentrations (25 nM and 50 nM) of siRNAs that target the coding region of UAP56 or control siRNA followed by infection with WSN at MOI 2 for 8h. (B) Total fluorescence intensity or nuclear to cytoplasmic fluorescence intensity (N/C ratio) (C) were quantified for images in A in which cells were treated with 25 nM siRNA targeting UAP56. For B (C, n = 117 cells; siRNA UAP56, n = 171 cells) and C (Control, n = 97 cells; siRNA UAP56, n = 166 cells). Graphs show data points and mean +/- SD. ****p<0.0001. (D) A549 cells were treated with 25 nM siRNA targeting UAP56 or control siRNA as in A. RNA-FISH was performed to detect poly(A) RNA. Total fluorescence intensity (E) or nuclear to cytoplasmic fluorescence intensity (F) were quantified for images in D. For (E) (C, n = 171 cells; siRNA UAP56, n = 213 cells) and F (Control, n = 172 cells; siRNA UAP56, n = 208 cells). Graphs show data points and mean +/- SD. ****p<0.0001. (G-J) A549 cells were treated with 1 nM or 20 nM siRNA targeting the 3’UTR of the UAP56 mRNA or with control siRNA and then infected with WSN at MOI 2 for 8h. (G) Purified RNA from total cell lysates was subjected to qPCR to measure UAP56 mRNA levels. (H) Cell lysates were also subjected to western blot to detect UAP56 protein and β-actin as control. Quantification of protein bands normalized to their loading control is shown at the bottom of the blots. (I) Purified RNA from total cell lysates in (G) were subjected to qPCR to measure viral mRNA levels. (J) Purified RNA from nuclear and cytoplasmic fractions from cells treated as in (G-J) were subjected to qPCR to measure viral mRNA levels in both fractions and determine their nuclear to cytoplasmic ratios (N/C). Control for cell fractionation is shown in S3 Fig. n = 3. Graphs are mean +/- SD. *p<0.05, **p<0.01, ****p<0.0001.
Fig 7.
Compound 2 activity phenocopies down-regulation of the mRNA export factor UAP56.
A549 cells or A549 cells stably expressing UAP56 E179A mutant were untreated or treated with control siRNA or with siRNA targeting the 3’UTR of UAP56 to knockdown endogenous UAP56 mRNA. Cells were then infected with WSN at MOI 2 for 8h followed by RNA-FISH to detect poly(A) RNA (A-C) or smRNA-FISH to detect M (D-F), HA (G-I), and NS (J-L) mRNAs. For B-C (C, n = 128 cells; UAP56-E197A+siRNA Control, n = 117 cells; UAP56-E197A_siUAP56-3’UTR, n = 170 cells). For E-F (C, n = 121 cells; UAP56-E197A+siRNA Control, n = 106 cells; UAP56-E197A_siUAP56-3’UTR, n = 108 cells). For H-I (C, n = 124 cells; UAP56-E197A+siRNA Control, n = 118 cells; UAP56-E197A_siUAP56-3’UTR, n = 151 cells). For K-L (C, n = 113 cells; UAP56-E197A+siRNA Control, n = 119 cells; UAP56-E197A_siUAP56-3’UTR, n = 115 cells). Graphs show data points and mean +/- SD. *p<0.05, **p<0.01, ***p<0.001 ****p<0.0001.
Fig 8.
Compound 2 alters the levels and intracellular distribution of a subset of cellular mRNAs.
Poly(A) RNA from total cell lysates, nuclear and cytoplasmic fractions untreated or treated with compound 2 was subjected to RNAseq analysis. Two biological duplicates were analyzed and the cut off is 1.5 fold for all analysis. RNAs selected were hits in both samples. (A) RNAs that are nuclear retained (yellow) or preferentially exported to the cytoplasm (light blue) are shown. Marked in red are RNAs whose total levels were not altered. Controls for fractionation are shown in S1 Table. (B) The number of RNAs that are up-regulated or down-regulated by compound 2 are shown. Marked in green are the number of RNAs known to be regulated by NS1 during infection. The identity of these RNAs are shown in S1 Table. (C-F) Selected mRNAs were also analyzed by qPCR to corroborate the RNAseq analysis. Relative mRNA levels and nuclear to cytoplasmic ratios of SPTLC3 (D), CEACAM19 (E), VTCN1 (F), and UQCC (G) were determined by qPCR from RNA obtained from total cell lysates, nuclear and cytoplasmic fractions treated with 0.1% DMSO or 2.5 μM compound 2 for 9 h. Three independent experiments were performed. C, control; Comp 2, Compound 2. Graphs show mean +/- SD. *p<0.05, **p<0.01, ***p<0.001 ****p<0.0001.
Fig 9.
Compound 2 inhibits viral protein production and replication.
(A) A549 cells were pre-treated with either 0.1% DMSO or 2.5μM compound 2 before infection with A/WSN33 at MOI 2 for 8 h. Cell lysates were subjected to western blot analysis to detect viral proteins including PB1, PB2, PA, NA, NS1, M1, M2, and HA. β-Actin was used as a loading control. This blot is a representative of three independent experiments. (B-D) Effect of compound 2 on cell viability and viral replication of (B) A/WSN/33 (H1N1), (C) A/Vietnam/1203/04 (H5N1), and (D) A/Panama/99 (H3N2) influenza A virus strains. Cell viability was determined by the MTT assay in cells treated for 24 h (H1N1 and H5N1) or 48 h (H3N2). Viral titer was determined by plaque assay in cells infected for 24 h (H1N1 and H5N1) or 48 h (H3N2) at MOI 0.01. Three independent experiments were performed. Error bars are SD.
Table 1.
HA mRNA Probes.